A mathematical procedure to estimate the impact of a change in method on discordance or misclassification at a decision limit in laboratory method comparison studies

Background: Laboratories often adopt new methods. It would be useful to have a statistical procedure to estimate the incremental impact of a change in assay.

Methods: Mathematical modeling, statistical analysis, and case example.

Results: We derived equations to estimate the proportion of discordant results that can be attributed to the new laboratory method. The calculations were demonstrated by comparing eGFR values based on creatinine values determined using the enzymatic method (existing method) and Jaffe method (new method). The discordance rate at the 60 ml/min eGFR decision limit was 3.15%. In this example, we estimated that 60% of the discordant results could be attributed to the Jaffe method.

Conclusion: The sources of discordance in a laboratory method comparison study can be divided into three categories: The baseline discordance due to imprecision in the established method, the incremental discordance due to imprecision in the new method, and lack of analytical specificity. Discordance due to imprecision can be attributed to each individual method. Discordance due to bias can be attributed to individual methods if information is available to estimate the rate of biased observations in either method. Such information can be used to estimate the incremental cost effectiveness associated with the adoption of a new method.